Just-in-time adaptive interventions

What are just-in-time adaptive interventions?

Just-in-time adaptive interventions (JITAIs) adjust support using information about a person’s changing state or context. Their aim is to provide suitable support at a suitable time, including withholding support when it would not help.

How JITAIs adapt support

Nahum-Shani and colleagues (2018) describe the components and design principles of JITAIs. They also identify a gap between technological capacity and evidence for effective development and evaluation.

Example of a just-in-time adaptive intervention

Consider a hypothetical support app with an agreed check-in each afternoon. That check-in is a decision point: a time to choose whether and how to respond. The user reports whether they are having a difficult moment and whether they are available for support. Those answers, together with their previously selected coping option, are tailoring information.

The intervention options might be a prompt for that coping option or no prompt. A simple decision rule could offer the prompt only when the person reports both difficulty and availability; otherwise, it offers nothing at that decision point. The check-in time can be fixed while the support delivered changes with the answers. This illustrates how the components fit together; it is not evidence that this rule selects the best timing or improves outcomes.

Designers distinguish vulnerability—a moment of increased need—from receptivity—whether the person can and wants to engage with support. A location sensor alone cannot establish both.

Evaluating timing and benefit

Sensor accuracy, intervention burden, privacy, and measured outcomes matter. Adaptive delivery and machine learning are not proof of benefit. The timing rule and support need evaluation in the intended population and setting.